Redesigning the physical layout of a village is simpler than transforming its social and productive life. For obvious reasons, political elites–parti… - James C. Scott

To grasp the prodigious variety of customary ways of measuring land, we would have to imagine literally scores of "maps" constructed along very different lines than mere surface area. I have in mind the sorts of maps devised to capture our attention with a kind of fun-house effect in which, say, the size of a country is made proportional to its population rather than its geographical size, with China and India looming menacingly over Russia, Brazil, and the United States, while Libya, Australia, and Greenland virtually disappear. These types of customary maps (for there would be a great many) would construct the landscape according to units of work and yield, type of soil, accessibility, and ability to provide subsistence, none of which would necessarily accord with surface area. The measurements are decidedly local, interested, contextual, and historically specific. What meets the subsistence needs of one family may not meet the subsistence needs of another. Factors such as local crop regimens, labor supply, agricultural technology, and weather ensure that the standards of evaluation vary from place to place and over time. Directly apprehended by the state, so many maps would represent a hopelessly bewildering welter of local standards. They definitely would not lend themselves to aggregation into a single statistical series that would allow state officials to make meaningful comparisons.

The subordination of metis is fairly obvious in the development of mass production in the factory. A comparable de-skilling process is, I believe, more compelling and, given the intractable obstacles to complete standardization, ultimately less successful in agricultural production.

A state mainly concerned with appropriation and control will find sedentary agriculture preferable to pastoralism or shifting agriculture. For the same reasons, such a state would generally prefer large-holding to small-holding and, in turn, plantation or collective agriculture to both. Where control and appropriation are the overriding considerations, only the last two forms offer direct control over the workforce and its income, the opportunity to select cropping patterns and techniques, and, finally, direct control over the production and profit of the enterprise. Although collectivization and plantation agriculture are seldom very efficient, they represent, as we have seen, the most legible and hence appropriable forms of agriculture.
The large capitalist agricultural producer faces the same problem as the factory owner: how to transform the essentially artisanal or metis knowledge of farmers into a standardized system that will allow him greater control over the work and its intensity. The plantation was one solution. In colonial countries, where able-bodied men were pressed into service as gang labor, the plantation represented a kind of private collectivization, inasmuch as it relied on the state for the extramarket sanctions necessary to control its labor force. More than one plantation sector has made up what it lacked in efficiency by using its political clout to secure subsidies, price supports, and monopoly privileges.